太阳辐射具有很强的非线性特征,给光伏发电并网带来诸多严重挑战。针对该问题,基于数值天气预报模式、机器学习和变分模态分解发展了一种订正预报方法：1）利用WRF-Solar模式对光伏站点的地表太阳辐射进行预报;2）采用变分模态分解（VMD）方法对其与观测值的偏差进行分解;3）利用双向循环神经网络（BiGRU）对分解后的各分量进行训练和预报;4）对各分量的预报进行求和后结合WRF-Solar的预报结果得到地表太阳辐射的订正预报结果。试验结果表明,经过VMD-BiGRU模型订正后,相比于WRF-Solar的预报结果MAE和RMSE的提升百分比分别为87.39%和87.29%,相关系数提高了0.25。
Abstract
Solar radiation has strong nonlinear characteristics, posing serious challenges to the grid integration of photovoltaic power generation. A correction forecast method has been developed, combining numerical weather prediction models, machine learning and variational mode decomposition, to address this issue： 1） Using the WRF-Solar model to forecast surface solar radiation at photovoltaic stations; 2） Using VMD decomposition method to decompose the bias between WFR-Solor forecast results and observation values; 3） Train and forecast the decomposed components using the BiGRU; 4） After summing the forecasts for each component, combined with the WRF-Solar results,the correction forecast of surface solar radiation is obtained. The experimental results show that after the VMD-BiGRU model correction, the percentage improvement of MAE and RMSE compared to WRF-Solar's forecast results is 87.39% and 87.29%, respectively, with a correlation coefficient increase of 0.25.
关键词
WRF-Solar模式 /
太阳辐射 /
机器学习 /
循环神经网络 /
变分模态分解
Key words
WRF-solar model /
solar radiation /
machine learning /
recurrent neural network /
variational mode decomposition
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 黄存瑞, 刘起勇. IPCC AR6报告解读: 气候变化与人类健康[J]. 气候变化研究进展, 2022, 18(4): 442-451.
HUANG C R, LIU Q Y.Interpretation of IPCC AR6 on climate change and human health[J]. Climate change research, 2022, 18(4): 442-451.
[2] IEA. Electricity Market Report 2023[R]. 2023.
[3] HUANG X Q, LI Q, TAI Y H, et al.Time series forecasting for hourly photovoltaic power using conditional generative adversarial network and Bi-LSTM[J]. Energy, 2022, 246: 123403.
[4] 申彦波, 赵宗慈, 石广玉. 地面太阳辐射的变化、影响因子及其可能的气候效应最新研究进展[J]. 地球科学进展, 2008, 23(9): 915-923.
SHEN Y B, ZHAO Z C, SHI G Y.The progress in variation of surface solar radiation, factors and probable climatic effects[J]. Advances in earth science, 2008, 23(9): 915-923.
[5] JIMENEZ P A, HACKER J P, DUDHIA J, et al.WRF-solar: description and clear-sky assessment of an augmented NWP model for solar power prediction[J]. Bulletin of the American Meteorological Society, 2016, 97(7): 1249-1264.
[6] 黄晶, 李锐海, 王颂, 等. 太阳辐射预报方法分类与评述[J]. 太阳能, 2023(3): 22-29.
HUANG J, LI R H, WANG S, et al.Classification and review of methods in solar radiation prediction[J]. Solar energy, 2023(3): 22-29.
[7] JAIHUNI M, BASAK J K, KHAN F, et al.A novel recurrent neural network approach in forecasting short term solar irradiance[J]. ISA transactions, 2022, 121: 63-74.
[8] DUAN J K, ZUO H C, BAI Y L, et al.Short-term wind speed forecasting using recurrent neural networks with error correction[J]. Energy, 2021, 217: 119397.
[9] 金存银, 张淑花, Xingong Li, 等. 地表太阳辐射短期预测方法研究进展[J]. 太阳能学报, 2023, 44(12):150-161.
JIN C Y, ZHANG S H, LI X G, et al.Research progress on short-term prediction methods of surface solar radiation[J]. Acta energiae solaris sinica, 2023, 44(12):150-161.
[10] 韩自奋, 颜鹏程, 李扬, 等. 基于短期历史资料的河西地区太阳辐射预报订正研究[J]. 干旱气象, 2022, 40(1): 125-134.
HAN Z F, YAN P C, LI Y, et al.Study on correction of solar radiation forecast in Hexi Region of Gansu Province based on short-term historical data[J]. Journal of arid meteorology, 2022, 40(1): 125-134.
[11] HAUPT S E, KOSOVIC B, JENSEN T, et al.The SunCast solar-power forecasting system: the results of the public-private-academic partnership to advance solar power forecasting[J]. National Center for Atmospheric Research(NCAR), Boulder (CO): Research Applications Laboratory, Weather Systems and Assessment Program(US), 2016.
[12] DRAGOMIRETSKIY K, ZOSSO D.Variational mode decomposition[J]. IEEE transactions on signal processing, 2014, 62(3): 531-544.
[13] BUSCEMA M.Back propagation neural networks[J]. Substance use & misuse, 1998, 33(2): 233-270.
[14] DEY R, SALEM F M.Gate-variants of gated recurrent unit(GRU) neural networks[C]//2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS). Boston, MA, USA, 2017: 1597-1600.
[15] 顾婷婷, 潘娅英, 张加易. 浙江省中尺度数值预报系统的地表太阳辐射预报订正方法[J]. 干旱气象, 2022, 40(2): 327-332.
GU T T, PAN Y Y, ZHANG J Y.Correction method of surface solar radiation forecast based on ZJWARMS[J]. Journal of arid meteorology, 2022, 40(2): 327-332.
[16] 达选芳, 李照荣, 王小勇, 等. 有云条件下太阳辐射短临预报订正技术研究[J]. 干旱气象, 2021, 39(6): 1006-1016.
DA X F, LI Z R, WANG X Y, et al.Correction technology of short-time solar radiation forecast based on cloud cover[J]. Journal of arid meteorology, 2021, 39(6): 1006-1016.
基金
甘肃省自然科学基金（22JR5RA481）; 兰州大学中央高校基本科研业务费专项（lzujbky-2021-sp61）
PDF(2796 KB)
